[EM] Double Defeat, Hare
Chris Benham
cbenhamau at yahoo.com.au
Mon May 19 09:05:16 PDT 2025
On April 9 last year I suggested the "Double Defeat, Hare" method
(nominating it as an alternative in a poll).
> *Voters strictly rank from the top however many candidates they wish and also may specify an approval cutoff.
>
> Default approval is only goes to top-ranked candidates.
>
> All candidates that are pairwise beaten by a more approved candidate are disqualified.
>
> If that leaves more than one qualified (i.e. not disqualified) candidate, commence eliminations according to Hare rules until only one qualified candidate remains.*
I now withdraw that not-so-clever idea because of this scenario:
43 A|
03 A<B|
44 B>C| (sincere is B| or B|>A or B>A|)
10 C|>A
A is both the normal Hare winner and the sincere Condorcet winner.
Approvals: C 54 B 47 A 46.
C>A 54-47, A>B 56-44, B>C 47-10.
Only A is disqualified by Double Defeat and C is eliminated by the
Hare rule leaving B the winner, rewarding the outrageous Burial strategists.
My favourite Condorcet method, Margins-Sorted Approval(explicit) would
punish them by electing C (as would the not-too-bad Smith//Approval).
It would first look at the BA pair because they are adjacent to each
other in the approval order and have a smaller margin of difference in
their approval scores (47-46=1) than the CB adjacent pair (54-47=7) and
notice that they are pairwise (by the rankings) out of order
and so flip that order, making it C>A>B. Neither of the two adjacent
pairs are now pairwise out of order so that order is final and C is on
top of it so C wins.
Smith//Approval(explicit) sees that all three candidates are in the
Smith set and so elects the one with the most approval.
Hare, being completely immune to Burial strategy (because it meets
Later-no-Help) elects (in this example) the sincere Condorcet Winner.
Chris Benham
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20250520/cf83b8b7/attachment.htm>
More information about the Election-Methods
mailing list